William Guthrie - "Second Shetland Truck System Report"

Lola Ridge - "Sun Up and Other Poems"

William Shakespeare - "A Midsumer Night's Dream"

Marietta Holley - "Samantha at Saratoga"

Species and Varieties, Their Origin by Mutation

H >> Hugo DeVries >> Species and Varieties, Their Origin by Mutation




Many of them have been tested as to their constancy from seed. Jordan
made numerous cultures, some of which lasted ten or twelve years; Thuret
has verified the assertion concerning their constancy by cultures
extending over seven years in some instances; Villars and de Bary made
numerous trials of shorter duration. All agree as to the main points.
The local races are uniform and come true from seed; the variability of
the species is not of a fluctuating, but of a polymorphous nature. A
given elementary species keeps within its limits and cannot vary beyond
them, but the whole group gives the impression of variability by its
wide range of distinct, but nearly allied forms.

The geographic distribution of these elementary species of the
whitlow-grass is quite distinct from that of the violets. Here
predominant species are limited to restricted localities. Most of them
occupy one or more departments of France, and in Holland two of them are
spread over several provinces. An important number are native in the
centre of Europe, and from the vicinity of Lyons, Jordan succeeded in
establishing about fifty elementary [50] species in his garden. In this
region they are crowded together and not rarely two or even more quite
distinct forms are observed to grow side by side on the same spot.
Farther away from this center they are more widely dispersed, each
holding its own in its habitat. In all, Jordan has distinguished about
two hundred species of _Draba verna_ from Europe and western Asia.
Subsequent authors have added new types to the already existing number
from time to time.

The constancy of these elementary species is directly proven by the
experiments quoted above, and moreover it may be deduced from the
uniformity of each type within its own domain. These are so large that
most of the localities are practically isolated from one another, and
must have been so for centuries. If the types were slowly changing such
localities would often, though of course not always, exhibit slighter
differences, and on the geographic limits of neighboring species
intermediates would be found. Such however, are not on record. Hence the
elementary species must be regarded as old and constant types.

The question naturally arises how these groups of nearly allied forms
may originally have been produced. Granting a common origin for all of
them, the changes may have been [51] simultaneous or successive.
According to the geographic distribution, the place of common origin
must probably be sought in the southern part of central Europe, perhaps
even in the vicinity of Lyons. Here we may assume that the old _Draba
verna_ has produced a host or a swarm of new types. Thence they must
have spread over Europe, but whether in doing so they have remained
constant, or whether some or many of them have repeatedly undergone
specific mutations, is of course unknown.

The main fact is, that such a small species as _Draba verna_ is not at
all a uniform type, but comprises over two hundred well distinguished
and constant forms.

It is readily granted that violets and whitlowgrasses are extreme
instances of systematic variability. Such great numbers of elementary
species are not often included in single species of the system. But the
numbers are of secondary importance, and the fact that systematic
species consist, as a rule, of more than one independent and constant
subspecies, retains its almost universal validity.

In some cases the systematic species are manifest groups, sharply
differentiated from one another. In other instances the groups of
elementary forms as they are shown by direct observation, have been
adjudged by many authors [52] to be too large to constitute species.
Hence the polymorphous genera, concerning the systematic subdivisions of
which hardly two authors agree. Brambles and roses are widely known
instances, but oaks, elms, apples, and pears, _Mentha_, _Prunu_s,
_Vitis_, _Lactuca_, _Cucumis_, _Cucurbita_ and numerous others are in
the same condition.

In some instances the existence of elementary species is so obvious,
that they have been described by taxonomists as systematic varieties or
even as good species. The primroses afford a widely known example.
Linnaeus called them _Primula veris_, and recognized three types as
pertaining to this species, but Jacquin and others have elevated these
subspecies to the full rank of species. They now bear the names of
_Primula elatior_ with larger, _P. officinalis_ with smaller flowers,
and _P. acaulis_. In the last named the common flower-stalk is lacking
and the flowers of the umbel seem to be borne in the arils of the basal
leaves.

In other genera such nearly allied species are more or less universally
recognized. _Galium Mollugo_ has been divided into _G. elatum_ with a
long and weak stem, and _G. erectum_ with shorter and erect stems;
_Cochlearia danica_, _anglica_ and _officinalis_ are so nearly allied as
to be hardly distinguishable. _Sagina apetala_ and _patula_, [53]
_Spergula media_ and _salina_ and many other pairs of allied species
have differentiating characters of the same value as those of the
elementary species of _Draba verna_. _Filago_, _Plantago_, _Carex_,
_Ficaria_ and a long series of other genera afford proofs of the same
close relation between smaller and larger groups of species. The
European frost-weeds or _Helianthemum_ include a group of species which
are so closely allied, that ordinary botanical descriptions are not
adequate to give any idea of their differentiating features. It is
almost impossible to determine them by means of the common analytical
keys. They have to be gathered from their various native localities and
cultivated side by side in the garden to bring out their differences.
Among the species of France, according to Jordan, _Helianthemum
polifolium_, _H. apenninum_, _H. pilosum_ and _H. pulverulentum_ are of
this character.

A species of cinquefoil, _Potentilla Tormentilla_, which is
distinguished by its quaternate flowers, occurs in Holland in two
distinct types, which have proved constant in my cultural experiments.
One of them has, broad petals, meeting together at the edges, and
constituting rounded saucer without breaks. The other has narrow petals,
which are strikingly separated from one another and show the sepals
between them. [54] In the same manner bluebells vary in the size and
shape of the corolla, which may be wide or narrow, bell-shaped or
conical, with the tips turned downwards, sidewards or backwards.

As a rule all of the more striking elementary types have been described
by local botanists under distinct specific names, while they are thrown
together into the larger systematic species by other authors, who study
the distribution of plants over larger portions of the world. Everything
depends on the point of view taken. Large floras require large species.
But the study of local floras yields the best results if the many forms
of the region are distinguished and described as completely as possible.
And the easiest way is to give to each of them a specific name. If two
or more elementary species are united in the same district, they are
often treated in this way, but if each region had its own type of some
given species, commonly the part is taken for the whole, and the sundry
forms are described under the same name, without further distinctions.

Of course these questions are all of a practical and conventional
nature, but involve the different methods in which different authors
deal with the same general fact. The fact is that systematic species are
compound groups, exactly like the genera and that their real units [55]
can only be recognized by comparative experimental studies.

Though the evidence already given might be esteemed to be sufficient for
our purpose, I should like to introduce a few more examples; two of them
pertain to American plants.

The Ipecac spurge or _Euphorbia Ipecacuanha_ occurs from Connecticut to
Florida, mainly near the coast, preferring dry and sandy soil. It is
often found by the roadsides. According to Britton and Brown's
"Illustrated Flora" it is glabrous or pubescent, with several or many
stems, ascending or nearly erect; with green or red leaves, which are
wonderfully variable in outline, from linear to orbicular, mostly
opposite, the upper sometimes whorled, the lower often alternate. The
glands of the involucres are elliptic or oblong, and even the seeds vary
in shape.

Such a wide range of variability evidently points to the existence of
some minor types. Dr. John Harshberger has made a study of those which
occur in the vicinity of Whitings in New Jersey. His types agree with
the description given above. Others were gathered by him at Brown's
Mills in the pinelands, New Jersey, where they grew in almost pure sand
in the bright sunlight. He observed still other differentiating
characters. The amount of seed [56] produced and the time of flowering
were variable to a remarkable degree.

Dr. Harshberger had the kindness to send me some dried specimens of the
most interesting of these types. They show that the peculiarities are
individual, and that each specimen has its own characters. It is very
probable that a comparative experimental study will prove the existence
of a large number of elementary species, differing in many points; they
will probably also show differences in the amount of the active chemical
substances, especially of emetine, which is usually recorded as present
in about 1%, but which will undoubtedly be found in larger quantities in
some, and in smaller quantities in other elementary species. In this way
the close and careful distinction of the really existing units might
perhaps prove of practical importance.

MacFarlane has studied the beach-plum or _Prunus maritima_, which is
abundant along the coast regions of the Eastern States from Virginia to
New Brunswick. It often covers areas from two to two hundred acres in
extent, sometimes to the exclusion of other plants. It is most prolific
on soft drifting sand near the sea or along the shore, where it may at
times be washed with ocean-spray. The fruit usually become ripe about
the middle of August, and show extreme [57] variations in size, shape,
color, taste, consistency and maturation period, indicating the
existence of separate races or elementary species, with widely differing
qualities. The earlier varieties begin to ripen from August 10 to 20,
and a continuous supply can be had till September 10, while a few good
varieties continue to ripen till September 20. But even late in October
some other types are still found maturing their fruits.

Exact studies were made of fruit and stone variations, and their
characteristics as to color, weight, size, shape and consistency were
fully described. Similar variations have been observed, as is well
known, in the cultivated plums. Fine blue-black fruits were seen on some
shrubs and purplish or yellow fruits on others. Some exhibit a firmer
texture and others a more watery pulp. Even the stones show differences
which are suggestive of distinct races.

Recently Mr. Luther Burbank of Santa Rosa, California, has made use of
the beach-plum to produce useful new varieties. He observed that it is a
very hardy species, and never fails to bear, growing under the most
trying conditions of dry and sandy, or of rocky and even of heavy soil.
The fruits of the wild shrubs are utterly worthless for anything but
preserving. [58] But by means of crossing with other species and
especially with the Japanese plums, the hardy qualities of the
beach-plum have been united with the size, flavor and other valuable
qualities of the fruit, and a group of new plums have been produced with
bright colors, ovoid and globular forms which are never flattened and
have no suture. The experiments were not finished, when I visited Mr.
Burbank in July, 1904, and still more startling improvements were said
to have been secured.

I may perhaps be allowed to avail myself of this opportunity to point
out a practical side of the study of elementary species. This always
appears whenever wild plants are subjected to cultivation, either in
order to reproduce them as pure strains, or to cross them with other
already cultivated species. The latter practice is as a rule made use of
whenever a wild species is found to be in possession of some quality
which is considered as desirable for the cultivated forms. In the case
of the beach-plum it is the hardiness and the great abundance of fruits
of the wild species which might profitably be combined with the
recognized qualities of the ordinary plums. Now it is manifest, that in
order to make crosses, distinct individual plants are to be chosen, and
that the variability of the wild species may be of very great
importance. [59] Among the range of elementary species those should be
used which not only possess the desired advantages in the highest
degree, but which promise the best results in other respects or their
earliest attainment. The fuller our knowledge of the elementary species
constituting the systematic groups, the easier and the more reliable
will be the choice for the breeder. Many Californian wild flowers with
bright colors seem to consist of large numbers of constant elementary
forms, as for instance, the lilies, godetias, eschscholtias and others.
They have been brought into cultivation many times, but the minutest
distinction of their elementary forms is required to attain the highest
success.

In concluding, I will point out a very interesting difficulty, which in
some cases impedes the clear understanding of elementary species. It is
the lack of self-fertilization. It occurs in widely distant families,
but has a special interest for us in two genera, which are generally
known as very polymorphous groups.

One of them is the hawkweed or _Hieracium_, and the other is the
dandelion or _Taraxacum officinale_. Hawkweeds are known as a genus in
which the delimitation of the species is almost impossible, Thousands of
forms may be cultivated side by side in botanical gardens, exhibiting
[60] slight but undoubted differentiating features, and reproduce
themselves truly by seed. Descriptions were formerly difficult and so
complicated that the ablest writers on this genus, Fries and Nageli are
said not to have been able to recognize the separate species by the
descriptions given by each other. Are these types to be considered as
elementary species, or only as individual differences? The decision of
course, would depend upon their behavior in cultures. Such tests have
been made by various experimenters. In the dandelion the bracts of the
involucre give the best characters. The inner ones may be linear or
linear-lanceolate, with or without appendages below the tip; the outer
ones may be similar and only shorter, or noticeably larger, erect,
spreading or even reflexed, and the color of the involucre may be a pure
green or glaucous; the leaves may be nearly entire or pinnatifid, or
sinuate-dentate, or very deeply runcinate-pinnatifid, or even pinnately
divided, the whole plant being more or less glabrous.

Raunkiaer, who has studied experimentally a dozen types from Denmark,
found them constant, but observed that some of them have no pollen at
all, while in others the pollen, though present, is impotent. It does
not germinate on the stigma, cannot produce the ordinary tube, [61] and
hence has no fertilizing power. But the young ovaries do not need such
fertilization. They are sufficient unto themselves. One may cut off all
the flowers of a head before the opening of the anthers, and leave the
ovaries untouched, and the head will ripen its seeds quite as well. The
same thing occurs in the hawkweeds. Here, therefore, we have no
fertilization and the extensive widening of the variability, which
generally accompanies this process is, of course, wanting. Only partial
or vegetative variability is present. Unfertilized eggs when developing
into embryos are equivalent to buds, separated from the parent-plant and
planted for themselves. They repeat both the specific and the individual
characters of the parent. In the case of the hawkweed and the dandelion
there is at present no means of distinguishing between these two
contrasting causes of variability. But like the garden varieties which
are always propagated in the vegetative way, their constancy and
uniformity are only apparent and afford no real indication of hereditary
qualities.

In addition to these and other exceptional cases, seed-cultures are
henceforth to be considered as the sole means of recognizing the really
existing systematic units of nature. All other groups, including
systematic species and [62] genera, are equally artificial or
conventional. In other words we may state "that current misconceptions
as to the extreme range of fluctuating variability of many native
species have generally arisen from a failure to recognize the composite
nature of the forms in question," as has been demonstrated by MacDougal
in the case of the common evening-primrose, _Oenothera biennis_. "It is
evident that to study the behavior of the characters of plants we must
have them in their simplest combinations; to investigate the origin and
movements of species we must deal with them singly and uncomplicated."


[63]

LECTURE III

ELEMENTARY SPECIES OF CULTIVATED PLANTS

Recalling the results of the last lecture, we see that the species of
the systematists are not in reality units, though in the ordinary course
of floristic studies they may, as a rule, seem to be so. In some cases
representatives of the same species from different countries or regions,
when compared with one another do not exactly agree. Many species of
ferns afford instances of this rule, and Lindley and other great
systematists have frequently been puzzled by the wide range of
differences between the individuals of a single species.

In other cases the differing forms are observed to grow near each other,
sometimes in neighboring provinces, sometimes in the same locality,
growing and flowering in mixtures of two or three or even more
elementary types. The violets exhibit widespread ancient types, from
which the local species may be taken to have arisen. The common
ancestors of the Whitlow-grasses are probably not to be found [64] among
existing forms, but numerous types are crowded together in the southern
part of central Europe and more thinly scattered elsewhere, even as far
as western Asia. There can be little doubt that their common origin is
to be sought in the center of their geographic distribution.

Numerous other cases exhibit smaller numbers of elementary units within
a systematic species; in fact purely uniform species seem to be
relatively rare. But with small numbers there are of course no
indications to be expected concerning their common origin or the
starting point of their distribution.

It is manifest that these experiences with wild species must find a
parallel among cultivated plants. Of course cultivated plants were
originally wild and must have come under the general law. Hence we may
conclude that when first observed and taken up by man, they must already
have consisted of sundry elementary subspecies. And we may confidently
assert that some must have been rich and others poor in such types.

Granting this state of things as the only probable one, we can easily
imagine what must have been the consequences. If a wild species had been
taken into cultivation only once, the cultivated form would have been a
single elementary [65] type. But it is not very likely that such
partiality would occur often. The conception that different tribes at
different times and in distant countries would have used the wild plants
of their native regions seems far more natural than that all should have
obtained plants for cultivation from the same source or locality. If
this theory may be relied upon, the origin of many of the more widely
cultivated agricultural plants must have been multiple, and the number
of the original elementary species of the cultivated types must have
been so much the larger, the more widely distributed and variable the
plants under consideration were before the first period of cultivation.

Further it would seem only natural to explain the wide variability of
many of our larger agricultural and horticultural stocks by such an
incipient multiformity of the species themselves. Through commercial
intercourse the various types might have become mixed so as to make it
quite impossible to point out the native localities for each of them.

Unfortunately historical evidence on this point is almost wholly
lacking. The differences in question could not have been appreciated at
that remote period, and interest the common observer but little even
today. The history of most of the cultivated plants is very obscure,
[66] and even the most skillful historians, by sifting the evidence
afforded by the older writers, and that obtained by comparative
linguistic investigations have been able to do little more than frame
the most general outline of the cultural history of the most common and
most widely used plants.

Some authors assume that cultivation itself might have been the
principal cause of variability, but it is not proved, nor even probable,
that cultivated plants are intrinsically more variable than their wild
prototypes. Appearances in this case are very deceptive. Of course
widely distributed plants are as a rule richer in subspecies than forms
with limited distribution, and the former must have had a better chance
to be taken into cultivation than the latter. In many cases, especially
with the more recent cultivated species, man has deliberately chosen
variable forms, because of their greater promise. Thirdly, wide
variability is the most efficient means of acclimatization, and only
species with many elementary units would have offered the adequate
material for introduction into new countries.

From this discussion it would seem that it is more reasonable to assert
that variability is one of the causes of the success of cultivation,
than to assume that cultivation is a cause of variability [67] at large.
And this assumption would be equally sufficient to explain the existing
conditions among cultivated plants.

Of course I do not pretend to say that cultivated plants should be
expected to be less variable than in the wild state, or that swarms of
elementary species might not be produced during cultivation quite as
well as before. However the chance of such an event, as is easily seen,
cannot be very great, and we shall have to be content with a few
examples of which the coconut is a notable one.

Leaving this general discussion of the subject, we may take up the
example of the beets. The sugar-beet is only one type from among a horde
of others, and though the origin of all the single types is not
historically known, the plant is frequently found in the wild state even
at the present time, and the native types may be compared with the
corresponding cultivated varieties.

The cultivation of beets for sugar is not of very ancient date. The
Romans knew the beets and used them as vegetables, both the roots and
the leaves. They distinguished a variety with white and one with red
flesh, but whether they cultivated them, or only collected them from
where they grew spontaneously, appears to be unknown.

[68] Beets are even now found in large quantities along the shores of
Italy. They prefer the vicinity of the sea, as do so many other members
of the beet family, and are not limited to Italy, but are found growing
elsewhere on the littoral of the Mediterranean, in the Canary Islands
and through Persia and Babylonia to India. In most of their native
localities they occur in great abundance.

The color of the foliage and the size of the roots are extremely
variable. Some have red leafstalks and veins, others a uniform red or
green foliage, some have red or white or yellow roots, or exhibit
alternating rings of a red and of a white tinge on cut surfaces. It
seems only natural to consider the white and the red, and even the
variegated types as distinct varieties, which in nature do not
transgress their limits nor change into one another. In a subsequent
lecture I will show that this at least is the rule with the
corresponding color-varieties in other genera.

The fleshiness or pulpiness of the roots is still more variable. Some
are as thick as the arm and edible, others are not thicker than a finger
and of a woody composition, and the structure of this woody variety is
very interesting. The sugar-beet consists, as is generally known, of
concentric layers of sugar-tissue and of vascular [69] strands; the
larger the first and the smaller the latter, the greater is, as a rule,
the average amount of sugar of the race. Through the kindness of the
late Mr. Rimpau, a well known German breeder of sugar-beet varieties, I
obtained specimens from seed of a native wild locality near Bukharest.
The plants produced quite woody roots, showing almost no sugar tissue at
all. Woody layers of strongly developed fibrovascular strands were seen
to be separated one from another only by very thin layers of
parenchymatous cells. Even the number of layers is variable; it was
observed to be five in my plants; but in larger roots double this number
and even more may easily be met with.

Copyright (c) 2007. topbookz.net. All rights reserved.